Brand safety is one of the hot topics in digital right now, and rightly so. Display spend and investment into video continues to rise, as advertisers and agencies continue to reap the benefits of programmatic.
While the focus has been audience led, and about harnessing the potential of audience segmentation and predictive analysis, it’s still key to assess the context of ad serving, and ensuring brands have a safe environment in which to advertise, and users have a positive experience with those brands.
This week, a number of large companies, including Havas (advertising agency) and M&S have pulled their display advertising spend from Google due to fears around brand safety and their ads being served alongside extremist content on the GDN and on YouTube.
Found share these concerns and, whilst we haven’t pulled any activity, we take our responsibility in this area very seriously. While the rise of new inventory is obviously a wonderful thing, it raises the risk that unwanted inventory will make its way through the safety net.
At Found, we work hard every day to combat this, training our analyst team and working with our platform partners to minimise the brand risks for our clients.
What measures are put in place?
- Digital Content Labels
Google classifies all online content into 5 broad labels. These are:
- General
- PG
- Teen and Older
- Mature
- Not yet labelled
The type of label selected will vary slightly from client to client, but at Found the majority of our campaigns will only target General, PG and Mature. Teen is often not relevant to our advertising campaigns and Not Yet Labelled is never used.
Indeed, Not Yet Labelled is always blocked from our campaigns. Google classification can take time, and given the volume of new sites, video and other content being uploaded each minute, there is a significant risk to utilising this label. Blocking ensures our clients’ brands aren’t exposed to the risk of appearing next to suspect content.
- Category Exclusions – Sensitive Content
Google breaks online content down into further categories for advertisers to target or exclude.
The biggest of these is sensitive content, which essentially covers any type of site that may be considered questionable, or could potentially cause offense. These include:
- Crime, police and emergency
- Death and tragedy
- Military and international conflict
- Juvenile, gross and bizarre
- Profanity and rough language
- Sexually suggestive
Best practise at Found is to block all of these to minimise risky exposure for our clients.
- Category Exclusions – Placements
There are additional targeting layers that could be used/blocked to ensure another layer of brand safety. Most of these cover placement types and include error pages, parked domains, gambling, in-game etc. As a best practise, we always exclude these too.
- Negative Keywords
We create account-level sets of negative keywords to ensure there is no suspect contextual matching occurring. This prevents our ads from serving next to risky content based on site content, video titles, descriptions etc.
These negatives cover the obvious categories, i.e. terrorism, pornography etc. but we always try to keep an open communication flow with our clients to ensure we add any terms that the brand has had negative exposure to in the past.
- Manual Placement review
While we can use all of the pre-launch targeting options to minimise the risk to brand safety, there are ways to monitor brand exposure.
Google is pretty transparent in terms of reporting and does allow advertisers to see what domains ads have been served on – they call this the Placement report.
On a typical GDN campaign, this Placement report will contain tens of thousands of placements so it can be a daunting document to review, however, we do ensure there is always a manual check in place so we can remove any sensitive sites.
The future?
Google has promised a swift and significant reply, citing a three-pronged approach that will place more emphasis on policies, controls and enforcement.
Given the huge loss of revenue for Google (hundreds of millions already) and the potential loss of market share to competitors (Bing, Facebook etc.) we expect huge leaps forward here.
Until then, we will continue to use every measure at our disposal to ensure that the advertising that we are responsible for is served in an appropriate manner and remains as relevant as possible to the audiences that we are targeting with it.
